Dr. Felix Hildebrand
Computational Material Science at the Intersection of Physical Modeling and Artificial Intelligence
"Research is formalized curiosity. It is poking and prying with a purpose." – Zora Neale Hurston
I am a research engineer working on computational material modeling on different scales. My current focus is on the intersection between physical modeling and artificial intelligence. The overall question in this process is how these two approaches can best be combined to "hybrid" methods. The goal is to substantially improve our products through faster and more precise models.
- PhD at the Institute of Applied Mechanics with Christian Miehe, University of Stuttgart
- Research assistant at the Center of Mechanics with Sanjay Govindjee, ETH Zurich
- Visiting student at the Department of Mechanical Engineering with Rohan Abeyaratne, Massachusetts Institute of Technology
M. Ganser et al. (2019)A finite strain electro-chemo-mechanical theory for ion transport with application to binary solid electrolytes
- M. Ganser, F.E. Hildebrand, M. Kamlah, R.M. McMeeking
- Journal of the Mechanics and Physics of Solids 125, 681-713
M. Ganser et al. (2019)An Extended Formulation of Butler-Volmer Electrochemical Reaction Kinetics Including the Influence of Mechanics
- M. Ganser, F.E. Hildebrand, M. Klinsmann, M. Hanauer, M. Kamlah, R.M. McMeeking
- Journal of The Electrochemical Society 166 (4), H167-H176
R.M. McMeeking et al. (2019)Metal Electrode Surfaces Can Roughen Despite the Constraint of a Stiff Electrolyte
- R.M. McMeeking, M. Ganser, M. Klinsmann, F.E. Hildebrand
- Journal of The Electrochemical Society 166(6) A984-A995
M. Mykhaylov et al. (2019)An elementary 1-dimensional model for a solid state lithium-ion battery with a single ion conductor electrolyte and a lithium metal negative electrode
- M. Mykhaylov, M. Ganser, M. Klinsmann, F.E. Hildebrand, I. Guz, R.M. McMeeking
- Journal of the Mechanics and Physics of Solids 123, 207-221
Interview with Dr. Felix Hildebrand
Research Scientist Computational Material Modeling
Please tell us what fascinates you most about research.
For me, research is about finding new solutions. On the path to these solutions, one has to deal with new, exciting and often surprising questions, take detours and face dead ends, and do all of this with a combination of curiosity, creativity and persistence. Research is never boring, you always keep on learning, meeting inspiring people and feeling the pulse of time. And it is a great satisfaction if the identified solutions end up improving one of our products.
What makes research done at Bosch so special?
For me, research at Bosch is special as it combines both width and depth: On the one hand, researchers from a large number of scientific disciplines work together on a broad spectrum of technically challenging products. Depending on the problem at hand, they form interdisciplinary teams. On the other hand, we do not hesitate to tackle large and very complex questions if they are decisive for the success of our products.
What research topics are you currently working on at Bosch?
The focus of my research is the question of how domain knowledge as material models, for example, can best be combined with artificial intelligence (AI) in the field of computational materials science. One related challenge is the automatic processing of materials science texts with the help of AI that takes advantage of domain knowledge. Another application is "hybrid" material models that combine AI-based and physically-based modeling approaches.
What are the biggest scientific challenges in your field of research?
Despite intense research in both the areas of computational material modeling and artificial intelligence and the constant huge progress in both fields, research at the intersection of both disciplines is relatively young. This includes the hybrid methods that combine extensive domain knowledge with AI. The big question is how such hybrid methods should be designed under given conditions to achieve an optimal synergy.
How do the results of your research become part of solutions "Invented for life"?
Methods such as hybrid material modeling have the overall goal of allowing faster and more accurate predictions of the behavior of existing as well as new, disruptive materials in our products. Such predictions substantially simplify and accelerate the design of innovative products with improved or completely new functionality and higher reliability. They thus enable us to provide new and better products to our customers much faster.